This spring, I will be teaching a practical 14-week course on machine learning.
On completion, you will be ready to use Machine Learning
algorithms at your job
or in a personal project.
Location: Shopzilla
Price: $50 per lecture, $600 for entire class
Date | Lecture Title | Topics Covered |
---|---|---|
March 22nd |
Introduction
Materials |
Overview of the class, Prerequisites for class, Application of machine learning, Defining learning models, Defining classifiers, Groundhog's Day, Testing a classifier, ROC, AUC, Precision, Recall, F-measure |
March 29th |
Linear Models
Materials |
Linear Algebra, Linear Regression, Logistic Regression, General Linear Models, Overfitting, Curve fitting |
April 12th May 17th (Recap class) |
Introduction to Discriminative Models
Materials |
Loss Functions, SVMs, SMO, Inner Products, Kernels, Gender classification |
April 19th |
Introduction to Probabilistic Models
Materials |
Bayes Theorem, Priors, Regularization, Naive Bayes, Mail Classification |
April 26th |
Clustering Algorithms
Materials |
Clustering, K-means, Spectral Clustering, Canopy Clustering, Agglomerative Clustering |
May 10th |
Graphical Models (part 1)
Materials |
Bayes Networks, Markov Networks, Factor models, Mixture Models, Hierarchical Models |
May 24th |
Graphical Models (part 2)
Materials |
Message Passing, Variational methods, MCMC, EM, Structured EM, Topic Modeling |
May 31st |
Nonparametric methods
Materials |
Density Estimation, Kernel Regression, KNN, Decision Trees, Splines, Chinese Restaurant Process, Gaussian Processes |
June 7th |
Discrete Sequential Prediction
Materials |
Hidden Markov Models, Forward-backward, Conditional Random Fields, Entity Extraction |
June 14th |
Time Series
Materials |
AR, VAR, ARMA, ARIMA, ARCH, GARCH, Box Jenkins, Outlier Detection |
June 21th |
Continuous Sequential Prediction
Materials |
Kalman filters, LDS, DBNs, Particle Filtering, Robot Control |
June 28th |
Online Learning
Materials |
Boosting, Regret and Mistake Bounds, Perceptron Algorithm, Passive Aggressive Algorithms, Vowpal Wabbit, Large-scale Learning Algorithms |
July 5th |
Unsupervised Learning
Materials |
Clustering, PCA, Robust PCA, MDS, JL Lemma, Random Projection, Image Compression |
July 12th |
Combining Learning Models
Materials |
Boosting, Bagging, Random Forests, Mixture Models, Model Averaging, Netflix Challenge |
These dates are tentative and may change to accommodate the schedules of attendees. Topics may also change to adjust to the tastes of students. Focus will be on applications. Further information for each lecture will be provided as needed and requested.
You will not need a single thing. No books are required. I will source supplementary readings from online material. You are not required to read them. All code will be executable on all OS platforms and consist of exclusively open-source software. As an aside, there are no homeworks. There are no tests. There are also no course credits granted.
This will be an in-person lecture. I will provide supplementary readings before each class, and lecture notes afterwards. There is no guarantee of videos of the lectures.